Multimodal large language models (MLLMs) are increasingly capable in medical imaging, yet most focus on single-image settings. Clinical interpretation often requires integrating evidence across multiple images, such as different modalities, views, or time points. However, large-scale medical multi-image data and traini...
Zhen Chen, Yihang Fu, Rong Zhou et al.· 0 citations
Generative AI chatbots built for mental health could extend access to care, but evidence from real-world use is limited. We report a single-arm, naturalistic pilot of a foundation model trained for mental health, among 299 US adults with at least moderate depressive or anxiety symptoms who were followed for up to 12 mo...
Thomas D. Hull, Lizhe Zhang, Caitlin A. Stamatis et al.· 0 citations
Large language models (LLMs) show promise in medical applications, but their ability to detect and correct errors in clinical texts remains under-evaluated, particularly beyond English. We introduce MedRECT, a bilingual benchmark for Japanese and English that formulates medical error handling as three subtasks: error d...
The facts a language model stores are tied to its parameter count, so small models that fit on edge devices fail on expert problems, which need specialized knowledge and follow multi-step procedures. Fine-tuning for a specific domain or task writes the knowledge into the parameters but must be repeated for every model...
Kenan Alkiek, David Jurgens, Vinod Vydiswaran· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Recent thinking models are capable of solving complex reasoning tasks by scaling test-time compute, but this scaling should be allocated in line with task difficulty. On one hand, short reasoning (underthinking) leads to errors on harder problems that require extended reasoning steps; but, excessively long reasoning (o...
Reliable model fingerprints are essential for protecting large language models (LLMs) against unauthorized redistribution and commercial misuse. In black-box deployment, verification is hindered by defensive filtering of suspected fingerprint queries, as well as by downstream model modifications that may weaken embedde...
Financial text embeddings must distinguish changes in event status, perspective, and obligations even when passages share similar wording. NMIXX adapts existing encoders through 18.8k source-linked triplets: paraphrases and Korean-English translations preserve meaning, while targeted financial rewrites introduce semant...
Hanwool Lee, Sara Yu, Yewon Hwang et al.· 0 citations
Interpreting language models remains challenging due to the existence of residual stream, which linearly mixes and duplicates features across adjacent layers, causing single-layer analyses to miss this cross-layer structure. Cross-layer sparse autoencoders (SAEs) address layer mixing but operate in continuous space, wh...
Ankur Garg, Xuemin Yu, Hassan Sajjad et al.· 0 citations
Language models (LMs) can "memorize" information, i.e., encode training data in their weights in such a way that inference-time queries can lead to verbatim regurgitation of that data. This ability to extract training data can be problematic, for example, when data are private or sensitive. In this work, we investigate...
Mansi Sakarvadia, Aswathy Ajith, Arham Khan et al.· 0 citations
Hindsight Experience Replay (HER) enhances sample efficiency in goal-conditioned reinforcement learning (RL) by relabelling failed trajectories with goals that were actually achieved. However, HER assumes access to a goal relabelling function and a predicate function that determines whether a goal has been satisfied. T...
Kevin Yandoka Denamgana\"i, Daniel Hernandez, Ozan Vardal et al.· 0 citations
We present CombEval, a dynamic benchmark for evaluating combinatorial counting in large language models. CombEval represents each problem as a typed Cofola specification over entities, combinatorial objects, object dependencies, and constraints, enabling controlled generation of natural-language counting problems with...
Yuxu Zhou, Ond\v{r}ej Ku\v{z}elka, Yuyi Wang et al.· 0 citations
The study of the Generalized Turing Test is initiated, a formal generalization of Turing's imitation game from humans to arbitrary interactive agents, yielding an inherently adaptive form of evaluation that does not rely on fixed benchmarks.
Daniel Mitropolsky, Riccardo Neumarker, Emanuele Rimoldi et al.· 0 citations
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.
MIT News · Artificial Intelligence· news.mit.eduAug 31, 2026
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.
